Research map / Detection and segmentation / 3D detection and driving
Task-Driven Implicit Representations for Automated Design of LiDAR Systems
Nikhil Behari, Aaron Young, Tzofi Klinghoffer, Akshat Dave, Ramesh Raskar
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use learned task-driven LiDAR designs instead of uniform ray sampling | smartphone flash-LiDAR · face scanning (Basel Face Model) | Chamfer distance / bandwidth: 15.2 -> 2.5 Mbps at 576 rays | Improved |
| use learned distributed designs instead of uniform sampling | distributed LiDAR · robot end-effector tracking | Frechet distance / bandwidth: 32.2 -> 20.8 Mbps at 1200 rays (2 sensors) | Improved |
| use motion-adaptive learned scanning instead of random/even sampling | scanning LiDAR · warehouse object detection | miss rate / bandwidth: 5.8 -> 0.6 Mbps at 18 rays | Improved |
| add ray-visibility (occlusion) term to target density (ablation) | implicit LiDAR density · end-effector tracking | Frechet distance | Improved |
| increase sensor count from 2 to 8 learned sensors (ablation) | distributed LiDAR · robot tracking, 1200 rays | bandwidth: 20.8 -> 14.0 Mbps | Improved |
| increase sensor count from 2 to 4 learned sensors (ablation) | face scanning LiDAR · face scanning, 576 rays | bandwidth: 2.5 -> 2.7 Mbps | No clear effect |
| replace co-located laser emitter with learned alternative emitter designs | fixed smartphone detector · emitter design | reconstruction quality | No clear effect |
| deploy learned 2/4/10-sensor designs on real single-photon LiDAR hardware | face scanning designs · real-world face scan | qualitative reconstruction | Improved |
Closest papers by meaning
- Towards Realistic Scene Generation with LiDAR Diffusion Models (CVPR 2024 · similarity 0.87)
- 3D LiDAR Mapping in Dynamic Environments using a 4D Implicit Neural Representation (CVPR 2024 · similarity 0.86)
- ICP-Flow: LiDAR Scene Flow Estimation with ICP (CVPR 2024 · similarity 0.86)
- LiFi: LiDAR Generation from Multi-View Images via Geometric and Semantic Collaborative Guidance (NeurIPS 2026 · similarity 0.86)
- LiDAR4D: Dynamic Neural Fields for Novel Space-time View LiDAR Synthesis (CVPR 2024 · similarity 0.85)
- Towards foundational LiDAR world models with efficient latent flow matching (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
